Project Summary Antimicrobial resistance is an increasing problem, and current drug pipelines are not keeping pace with the rise of antimicrobial resistance. An alternative strategy is to boost host immunity. An often overlooked side-effect of the vitamin and mineral supplementation projects of the 1940s is that these supplements greatly reduced infectious disease burden. Recent work has shown that further gains may be possible, especially in adding phytochemicals back to highly processed diets typically consumed in the United States. However, we lack a fundamental understanding of how these components are processed by the microbiome, and how diet-derived molecules, microbiome and host immune system work together to resist infectious disease. A key barrier preventing us from making these discoveries is that each individual assay (microbiome, host gene expression metabolome, dietary compounds) is expensive and highly multivariate. Three key insights that enable the current project are the miniaturization of DNA and RNA sequencing assays on advanced nanoliter- scale liquid handling robots, greatly reducing the cost, the combination of untargeted and targeted mass spectrometry on the same samples in high throughput to enable discovery of a much greater chemical space, and the ability to use explicitly spatial maps on multiple scales to integrate the dataset throughout the body and enable both visual analytics and deep learning approaches based on spatial data. These breakthroughs will provide a fundamentally new understanding of how dietary metabolites promote disease resistance, and will allow us to develop a new infrastructure to integrate results from many investigators in different laboratories studying various aspects of these systems. Additionally, the results will allow us to choose biomaterials and biomarkers in human subjects that provide maximum information about internal nutritional and immune status. Results will be tested against the NHANES and American Gut cohorts. The results of this project will therefore be: 3D maps of mouse models showing how the microbiome, diet, and host gene expression produce immunity; an infrastructure for creating and sharing these maps; and a preliminary test of whether the results extend to large human populations.